Ali Bereyhi
PhD (Dr.-Ing.) | Senior Member IEEE
Assistant Professor (TS)
Department of Electrical and Computer Engineering
University of Toronto
Faculty Affiliate at Schwartz Reisman Institute (SRI)

Bahen Centre for Information Technology | BA7208
Email: ali(dot)bereyhi(at)utoronto(dot)ca

About Me

I am Assistant Professor - Teaching Stream in the Department of Electrical and Computer Engineering at the University of Toronto. I am also Faculty Affiliate at Schwartz Reisman Institute (SRI). From 2023 to 2024, I was Postdoctoral Research Fellow at Wireless Computing Lab (WCL) in the University of Toronto. From 2020 to 2023, I was Postdoctoral Research Associate and Lecturer at the Institute for Digital Communications (IDC) in the Friedrich-Alexander-Universität Erlangen-Nürnberg (FAU), Germany. I have received my PhD (in German Doktor der Ingenieurwissenschaft) with distinction from FAU in 2020, supervised by Prof. Dr.-Ing. Ralf R. Müller and Prof. Dr. Hermann Schulz-Baldes.

My main area of expertise is Machine Learning with focus on Deep Learning and its applications to Communications and Signal Processing. My research also investigates the connections between Theoretical Physics and the Theory of Learning. This has been the topic of my doctoral dissertation, entitled Statistical Mechanics of Regularized Least Squares, and is still one of my research areas.

News

[May 2025] I am teaching a new graduate course on Deep Generative Models. The course page is public and you are more than welcome to read the lecture notes, follow course materials, and watch the videos there. Quercus page is also left open to all UofT people, though all course materials are shared over the public page of the course. This is the first course lunched in the project <bereyhi-courses> in which I plan to unify all my courses for better access and maintenance.
[April 2025] Our paper "Universal Training of Neural Networks to Achieve Bayes Optimal Classification Accuracy" won the Best Student Paper Award be presented at ICASSP 2025
[Jan 2025] The lecture pages of the Courses ECE 1513 and ECE 1508 are now open to all UofT affiliates
[Jan 2025] Check out our recent work on "Universal Training of Neural Networks to Achieve Bayes Optimal Classification Accuracy" which is to be presented at ICASSP 2025
[Oct 2024] I am teaching "ECE 1508: Applied Learning" and "ECE 1513: Introduction to Machine Learning" in Winter 2025.
[Sep 2024] I am attending MLSP 2024, this year at Imperial College.
[Sep 2024] I am teaching "ECE 1508: Applied Learning" in Fall Semester 2024.
[May 2024] I am giving the course "ECE 1508 - Special Topics in Communications: Reinforcement Learning" in Summer Semester 2024.
[April 2024] The page for the course "Applied Deep Learning" has been migrated permanently to this link. To watch the course videos you need to login with UTORid. The lecture-notes are however available publicly.
[January 2024] I am giving the course "ECE 1508 - Special Topics in Communications: Applied Deep Learning" in Winter Semester 2024.
[January 2024] I am starting my new position as Assistant Professor - Teaching Stream in the Department of Electrical and Computer Engineering at the University of Toronto.
[September 2023] My grant proposal entitled “Bayesian Learning and Model Fitting via Nonlinear Models” has won the German Research Foundation (DFG) Walter-Benjamin Postdoctoral Fellowship Award hosted by Princeton University.
[June 2023] I am now on the Mathematics Genealogy Project. Check out my page here!
[May 2023] My course "Information Theory and Coding" in Winter 2022-23 has been selected as the Best Lecture in the Faculty of Engineering at FAU.
[Feb 2023] I am joining the Wireless Computing Lab (WCL) in the Department of Electrical and Computer Engineering at the University of Toronto.
[Nov 2022] The book "Compressed Sensing in Information Processing" has been published. This book has been prepared throughout the last phase of the DFG-funded special session Compressed Sensing in Information Processing (CoSIP). Checkout our contribution in Chapter 5. In this chapter, we have given a comprehensive introduction to the replica method and its application to asymptotic analysis of Bayesian-type algorithms.
[Jul 2022] I am giving an invited talk at the Random Matrices and Random Landscapes (RMRL 2022) conference in honor of Yan Fyodorov's 60th birthday. The talk is entitled "Bayesian Inference with Nonlinear Generative Models". The slides of my presentation can be found here.

Research Interests

Machine Learning and Statistical Learning Theory
Bayesian Inference, Machine Learning, Regression, Approximate Message Passing, Federated Learning
Statistical Signal Processing
Compressive Sensing, Detection Theory, Estimation Theory, Sensor Fusion
Statistical Mechanics and Its Connection to Information Processing
Replica Method, Information-theoretic Interpretation of Physical Models, Large Deviations Theory, Mean-field Theory, Random Matrix Theory
Information Theory
Network Information Theory, Information Theoretic Secrecy, Secret Key Agreement
Computation and Communications
Computation Coding, Distributed Coded Computing
Wireless Communications
Signal Processing for Communications

Potential Graduate Projects

Those students who are interested in graduate research projects either in the form of Research Internship or Graduate Course ECE2500 may reach out through email to schedule a meeting. For particular openings, I may also post some separate advertisement.

[Jan 2025] I have currently a few openings for ECE2500. The topics lies within the intersection of Wireless Communications and Deep Learning.

For Post-graduate Applicants

Unfortunately, I am not able to respond to most of application emails. If you don't hear back from me in a course of one week or so, it is safe to consider the case closed. In that case, please accept my sincere apology for not being responsive to your email. I wish you all success with your application and thank you for taking time to write to me.

For My Former Students at FAU

The lecture materials of my courses at FAU can still be accessed through StudOn. You can also download the lecture-notes here. The lecture-notes have been updated last time on Winter 2022-23. The version on GitHub is the final version, and hence could contain some changes compared to the older versions on the StudOn pages of previous semesters.

Page updated

Google Sites

Report abuse